AN ACCURATE METHOD FOR THE DETERMINATION OF BLOOD … · an accurate method for the determination...

AN ACCURATE METHOD FOR THE DETERMINATION OF BLOOD UREA NITROGEN BY DIRECT

NESSLERIZATION

BY CLEON J. GENTZKOW*

WITH THE ASSIST~XCE OF JOHN M. M.&sENt

(Prom the Division of Chemistry and Physics, Army Medical School, Washington)

(Received for publication, February 16, 1942)

The multiplicity of published procedures for det,ermining urea nitrogen in blood is an indication that the ideal method has not yet been found. That of Van Slyke and Cullen (1) is accurate but requires 3 cc. of whole blood, involves aeration, and is time-consuming. Of equal accuracy is the gasometric method of Van Slyke (2), but special apparatus and skill, not available in the ordinary clinical laboratory, are required.

Because of these factors, many simpler methods have been devised, usually at the expense of accuracy. In general they depend upon direct nesslerization, with or without the use of protective colloids, of blood filtrates in which the urea has been hydrolyzed by one means or another. While it is recognized by all that the accuracy of these simplified methods is not great, their use has been justified on the grounds that the error is not clinically significant. Of this type are the procedures proposed by Karr (3), Barrett (4), Raices (5), Wrenn (B), Taylor, Hayes, and Wells (7)) Hawk and Andes (8)) Naumann (9)) Hoffman and Osgood (lo), Ohllson (ll), Sure and Wilder (12), and Hughes and Saifer (13). With reference to these methods, Folin (14) has stated:

“It is unfortunate that accurate urea determinations cannot be made by direct Nesslerization of hydrolyzed blood filtrates. The reason why this cannot be done is not so much due to the insidious turbidities, for these can be largely eliminated by the use of gum ghatti, but rather because of the presence of relatively large amounts of nitrogenous materials which have a profound effect both on the quality and on the quant,ity of color which the urea-ammonia gives with Nessler’s reagent.”

The principal sources of error in the direct nesslerization methods, there- fore, are turbidity and the difference in quality of color in specimen and standard.

Turbidity is apparently due to the reduction of the complex mercuric salt of Nessler’s reagent to mercurous compounds by certain constituents normally present in blood, such as creatine, creatinine, uric acid, glucose,

* Lieutenant Colonel, Medical Corps, Unit.ed States Army. Chief of Division of Chemistry and Physics.

t Master Sergeant, Medical Department, United States Army.

531

by guest on April 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

532 DETERMINATION OF BLOOD UREA NITROGEN

etc. These reducing substances cause the development of a greenish precipitate in directly nesslerized specimens on prolonged standing.

To prevent this turbidity, Looney (15) introduced the use of gum ghatti, but we have found that the prevention is more apparent than real and depends on colloidal dispersion by the gum giving the appearance of a clear solution. Upon standing, the same greenish precipitate collects in about the same time as in the unprotected solution. Furthermore, the addition of the gum decreases the intensity of color by as much as 10 per cent (see Table I).

One of the principal complaints of those using a direct nesslerization method is that the shade of color in the unknown differs so markedly from that of the standard that when matching is done in a visual colorim- eter a certain amount of guesswork is involved. The remark is frequently heard that the filtrate appears to be less red than the standard.

Spectrophotometric study should show the reason for this difference in the quality of color, but nowhere in the literature is there any reference to such work, probably because of the instability of directly nesslerized specimens (Table I).

In the method to be described, we have stabilized the color and clarity of directly nesslerized blood filtrates for a period of at least 1 to 3 hours. This permitted the establishment of a narrow band of wave-lengths over which color absorption is approximately the same for directly nesslerized specimens and for ammonium sulfate standards.

EXPERIMENTAL

Early in this study it became apparent that methods previously used as standards against which to compare direct nesslerization procedures would not be sufficiently accurate and reproducible for our purposes. It was necessary to have for reference a method which could be used colori metrically to check the quantitative precision of direct nesslerization procedures, and one which would serve also as a basis for comparison of the quality of the resulting colors.

In the procedure of Van Slyke and Cullen, 3 cc. of treated whole blood are aerat,ed and the collected ammonia determined by titration, with a negative error of 1 to 2 per cent, due to incomplete recovery of the ammonia. If this method is applied to 5 cc. of blood filtrate (0.5 cc. of blood) and the ammonia loss is of the same magnitude, the expected error will be about 10 to 12 per cent. When ammonium sulfate solutions, equivalent in nitrogen content to 0.5 cc. of blood, and blanks were run by the aeration method, the collected ammonia nesslerized, and the blank subtracted, the error was found to be exactly of this order. If the blank was ignored, closer agreement resulted. The reduction of Nessler’s reagent in the blank ap-


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

C. J. GENTZKOW 533

pears to be due to substances from the rubber tubing connections which persist in spite of the most thorough cleansing.

Somewhat similar findings were obtained when the simple distillation method of Folin and Svedberg (16) was used. Here again fairly accurate figures resulted if no blank determination was made, but if the blank was deducted, standard solutions gave an error of about - 10 per cent. The blank in this method was found to be*due to the oil used as an antifoaming agent as well as to the rubber tubing connections. Compensating for this addition of substances reacting with Nessler’s reagent is the loss which occurs during the initial stages of the distillation in which about 90 per cent of the ammonia comes over. No matter how carefully one attempts to regulate the displacement of air from the apparatus, it bubbles through the receiving acid rapidly and vigorously, carrying ammonia with it and preventing complete absorption as demonstrated by Miller (17). Only when the all-glass steam distillation micro-Kjeldahl apparatus of Kemmerer and Hallett (18) was used could the nitrogen from an ammonium sulfate solution equivalent in amount to that in 0.5 cc. of normal blood (0.05 to 0.075 mg. of Nz) be recovered with an accuracy of 1 per cent.

New Distillation Method-The reference method finally adopted, in which this apparatus is used, is based on that of Folin and Svedberg and upon the semimicro-Kjeldahl method of Redemann (19). The flask and condenser are thoroughly swept out with steam from the generator and 1 cc. of N sulfuric acid plus 10 cc. of water is placed in the distillation flask. After the hydrolyzed blood filtrate or the known ammonium sulfate solution is introduced into the distillation flask, the burner under it is started. Heating of the generator and distillation flask is continued until all air has been swept from the apparatus, indicated by cessation of bubbling through the receiver. Alkali, in this instance saturated sodium borate solution, is then added to the specimen, liberating the ammonia. Since all air has been removed, there is no possibility of loss of ammonia by dilution with it or by too vigorous bubbling through the receiving acid. Distillation is continued in the usual manner.

This method was used in all of the succeeding work. Standard solutions and blanks were distilled in duplicate with each blood filtrate and only when the results agreed within 1 per cent were they used for comparison with the corresponding direct nesslerixations.

Stabilization of Color of Nesslerized Blood Filtrates; Use of Wetting Agents-The initial clarification of nesslerized solutions obtained by Looney by the use of gum ghatti was probably due to its dispersing properties. In the hope of obtaining a more permanent dispersion, a number of wetting agents were tried. Among them were Triton NE, Santomerse No. 1, Santomerse No. 2, Avitex W, Aerosol OT, Nacconol NR, Duponol


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


ME, and Soricin (sodium ricinoleate). In general these agents were similar in their action to gum ghatti. Nacconol NR was definitely superior to the others. When combined with potassium persulfate their action was somewhat more efficient, but not sufficiently so to warrant inclusion in the method finally adopted.

Use of Oxidants-Since the turbidity which develops is probably due to the reduction of Nessler’s reagent with the formation of insoluble mercurous compounds, the use of oxidizing agents suggested itself. Such an agent should have an oxidation-reduction potential sufficiently high to prevent reduction of the Nessler’s reagent without at the same time oxidizing the ammonium salts, the other nitrogenous constituents of the blood filtrate, or the Nessler’s reagent itself.

A large number of oxidizing agents, including persulfates, peroxides, hypochlorites, chlorates, and perchloric acid, were tried for this purpose. All had some effect in preventing turbidity, potassium persulfate being by far the best. However, after its addition color density progressively increased. Eventually a reddish sediment of mercuric iodide formed. This oxidation of the Nessler’s reagent indicated that the oxidation-reduction potential of the potassium persulfate was too high for the purpose.

In an attempt to restrain this action of the potassium persulfate to a point where oxidation of the mercuric complex would not occur, salts of various organic acids having mild reducing properties were added. A number of compounds were tried, including lactates, citrates, maleates, phthalates, tartrates, and gluconates. The desired results were obtained with the last two, the gluconates being somewhat better than the tartrates. When potassium persulfate and potassium gluconate were added to directly nesslerized blood filtrates, the specimens remained clear, with no change in color density for a minimum period of 1 hour to a maximum of 3 hours. The color int’ensity then gradually increased at about 1 per cent per hour. These final reaction mixtures do, however, remain free from turbidity for days.

This stabilization of reaction may be due to the attainment of a proper oxidation-reduction potential by the addition of gluconate and persulfate or it may be due to the formation of a soluble mercury complex by the gluconate, just as tartrate in an alkaline copper solution forms a soluble complex. Potassium gluconate alone has considerable stabilizing action but eventually the Nessler’s reagent is reduced and a greenish sediment forms. In all likelihood the action is a combination of both of these factors.

Analysis of Color Quality in Nesslerixed Xolutions-While the addition of persulfate and gluconate resulted in a clear reaction mixture which remained stable in color intensity for as long as 3 hours, it did not affect the


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

C. J. GENTZKOW 535

difference in shade of color between nesslerieed hydrolyzed blood filtrates on the one hand and similarly nesslerized ammonium sulfate standards and distilled blood filtrate on the other. The increased stability of color and clarity of solution, however, did permit spectrophotometric study. From the curves so obtained and shown in Fig. 1, the reason for the difference in shade becomes apparent.

go-

80-

70-

60-

50 -

40-

30-

- OISTIL LEO -NESSLERlIEO

---.OIRECTLY -NESSLERIZEO

I, 7 ---(NN,)2 so4 STANDARD

lo’,

\

0’~‘1~I,‘.~~l~l~I.l~I 420 440 460 480 500 520 ,540 560 580 600

WAVE-.h~fGTfi IN MLLI~WICR~NS

I?IG. 1. Spectral transmittance of a directly nesslerized blood filtrate after urcase hydrolysis compared with the same specimen in which the ammonia was separated by distillation and nesslerized. Concentration of urea N, 13.6 mg. per 100 cc. in both when compared with a standard at 500 rnp. A nesslerized ammonium sulfate solution equivalent to 15 mg. of X per 100 cc. is shown for comparison. If the ammonium sulfate solution were made exactly the same in concentration as the distilled specimen, their curves would be exactly superimposed.

By means of a Coleman photoelectric spectrophotometer, model 10-S, with fixed 5 rnp slit, the spectral transmittance of a directly nesslerized specimen was measured and compared with that of a nesslerized distillate of the same filtrate, as well as with a nesslerized ammonium sulfate solution of the same ammonia content. Repeated measurements of this kind show that transmission curves of distilled and nesslerized blood filtrates are


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


superimposable upon those of equimolar nesslerized ammonium sulfate standards. This is evident in Fig. 1 in which the similarity of color content in the two solutions is shown by the parallelism existing between them when their concentrations differ slightly.

This identity of curves, however, does not hold for an equivalent amount of ammonia in a directly nesslerized blood filtrate. Only between the wave-lengths of 490 and 510 rnp does the color density of the directly nesslerized specimen coincide with that of the distilled nesslerized filtrates and the standard solutions. Below this narrow band, the directly nesslerized filtrates transmit less light than do standards of equivalent ammonia concentration. Accordingly, calculations on unknowns, based on comparison with standard solutions at such wave-lengths, yield ammonia nitrogen values which are too high. Above this narrow band, the transmission of the directly nesslerized filtrates is greater than that of the standard solutions, and calculations based on the comparative color densities at these wave-lengths yield values which are too low.

Comparisons have been made by other methods at wave-lengths below 500 rnp. For example, a dominant wave-length of 440 mp was used by Sure and Wilder and Hughes and Saifer, and one of 430 rnp by Hoffman and Osgood. That such measurements have an inherent error is evident from the graphs in Fig. 1. The curve of the standard solution represents a concentration of 15 mg. of urea N per 100 cc. of blood. By calculation against this standard as read at 500 rnp, the concentration of the unknown, distilled as well as directly nesslerized, is 13.6 mg. per 100 cc. If, however, the comparison is made at 440 rnp, the values are 13.8 and 16.76 mg., and at 430 rnp, 13.8 and 17.0 mg. per 100 cc., respectively.

An idea of the magnitude of the errors resulting when unknown directly nesslerized blood filtrates are compared with standards at wave-lengths other than 500 rnp also may be obtained by inspection of Fig. 2. The three examples given are typical of the large number of specimens varying in concentration from 6 to 75 mg. of urea N per 100 cc. that have been studied by this method.

In each case, the color densities of directly nesslerized and distilled nesslerized blood filtrates were measured at intervals over the range of the visible spectrum. The calculation of the concentration of the unknown was based upon the previously noted finding that the transmission of distilled nesslerized filtrates was identical with that of ammonia standards of equivalent concentration. Plotting the concentration of nitrogen in directly nesslerized filtrates against wave-length emphasizes the great variation in results which occurs if the comparison with standards be made at wave-lengths other than those centering at 500 mp.

The values shown in Fig. 2, A were obtained with a Bausch and Lomb


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

C. J. GENTZKOW 537

I”“““’ “‘1”“‘1 ”

UREA NITROGEN CONC.

I%0 t

MG. Np/‘oo CC.

L

/5.0-

13.0 -

/LO-

9fo-

7.0 -

5.0 -

3.0 -

t.0 -

--__ A. VISUAL SPECTROPHOTOMETER

--._ --._

--._ --s- _ ---------.---.---- ------.-----.

-.__ 4

I

--..._ -.

--- -“‘.*. --..

. - - - _ _ .

___--- DIRECTLY -NESSLERUEO

I ' 1 3 1 ' ' 1 ' f L -.-.. OlSTlL L ED - NESSL ERIZEO

- (NH& SO4 STANDARD

lZO_ A 5 .PUOTOELECTf?tC COLOR/METER _ --.. ‘g --“_

f50- ---___

--_ -.- L.--__ A 6

--. .-...- .-.C-- .-.-.-. ___,

13.0 - EVELYN FILTERS c -..__ 0 E F G 77. -,-

.-. A. - NO. 420 Op... ‘7. -NO. 440

Il.O- C. - NO. 490

4

---.. E l.

D. - NO. 520 “.. E. - NO. 540 f ,\\\

P.O- f.- NO. 56.5 G. - NO. 580 ;--.. H 1

H. - NO. 600 -- ._____-,

I. - NO. 620 zo 4 14 I I I I I I 3 I I I,,,, I ,,

420 440 460 460 500 520 540 560 580 600

WAVE-LENGTH lN ~!ILLIMIcRONS

FIG. 2. A, the urea nitrogen concentration of two blood specimens as determined by the described method of direct nesslerization is compared with that of the same specimen in which the ammonia was separated by distillation and nesslerized. The readings were made on a Bausch and Lomb visual spectrophotometer. B, a comparison is made of the urea nitrogen of a single blood specimen as determined by direct nesslerization and by distillation and nesslerization. The latter is compared with an ammonium sulfate standard. The readings were made on an Evelynphoto- electric calorimeter with the filters designated.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


visual spectrophotometer and those of Fig. 2, B with the standard Evelyn photoelectric calorimeter in which the wave bands were limited by the use of glass filters. In the latter, further verification of the relationships dis- cussed under Fig. 1 is shown by the exact parallelism which results when the calculated concentration of a distilled nesslerized sample and that of a standard solution are plotted against wave-length.

Although the wave-length at which absorption is the same for both directly nesslerized and distilled nesslerized specimens is not always exactly at 500 rnp, it is never below 490 nor above 510 mp.

From the foregoing presentation, it is obvious that no application of Beer’s law to the color of a nesslerized blood filtrate at wave-lengths other than those centering at 500 rnp can give correct values when the standard of comparison is a known solution of ammonium sulfate. When a spectrophotometer is used, narrow wave-lengths in this region are easily obtained. When, however, a photoelectric calorimeter or a visual instrument is used, narrowness of wave band is secured by means of a proper light filter. Selection of such a filter involves a certain amount of empiricism, since monochromatic filters are neither available nor practicable. Inasmuch as filter density increases with selectivity, some sacrifice of narrowness of band must be made in order to avoid a density too great to permit sufficient light to pass.

The filter best suited in photoelectric calorimetry is that supplied with the Evelyn instrument under No. 49O.l It is equally satisfactory with other photoelectric instruments having sufficient sensitivity to permit full scale deflection when inserted in the light path. It may be made by assembling in a suitable holder Corning Glass Works standard glasses as follows: No. 338 Noviol shade C 2 mm. thick, No. 430 dark shade blue- green 2 mm. thick, and No. 503 dark theater blue 5 mm. thick.2

For visual calorimetry the most suitable one was found to be Wratten gelatin filter No. 75. Visibility with this filter is fairly good and the wave- lengths transmitted below 490 rnp are about balanced by those above 500 mp. This filter may be cemented between circular microscopic cover-slips and used in the eyepiece of the calorimeter, or between microscopic slides and used over the light source in the calorimeter base. Sand-blasting one surface of one of the slides improves diffusion. In either event the blue daylight filter must be removed, since it will alter the balance between unwanted wave-lengths.

1 The calorimeter and filters were manufactured by the Rubicon Company, Phila- delphia.

2 The thickness of this element may be reduced to 4 mm. for calorimeters having filter disks whose total capacity is only 8 mm. The effective transmission is not changed too greatly thereby.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

C. J. GENTZKOW 539

Method

Reagents- Sodium tungstate, 10 per cent aqueous solution. Lfh~lfuric acid, 0.66 N.

Stock standard nitrogen solution. Dissolve 7.074 gm. of ammonium sulfate, A. C. S., in 0.1 N sulfuric acid and make up to 1 liter.

Working standard. Dilute 5 cc. of stock solution to 500 cc. with 0.01 N sulfuric acid (0.015 mg. of N per cc.).

Potassium gluconate, 1 per cent aqueous solution. Store in the refrigerator. Make up a fresh solution weekly.

Potassium persulfate. A 2.5 per cent aqueous solution of the C.P.

nitrogen-free salt. This solution must be prepared freshly each week and kept in the refrigerator at all times. During warm weather remove it only long enough to measure out the amount required, for decomposition is quite rapid at higher temperatures.

Urease, Squibb, double strength, powdered. Nessler’s solution. The stock solution is the original one of Folin and

Wu (20)) metallic mercury being used (see also (14) p. 337). This Nessler’s reagent is superior to that of Koch and McMeekin (21). The final solution differs in that the alkali concentration is 5 instead of 7 per cent. Place 500 cc. of 10 per cent NaOH in a liter volumetric flask, add 150 cc. of stock Nessler’s reagent, and dilute to the mark with distilled water. Allow to stand for 3 days before use to permit sedimentation of the small amount of precipitate which forms.

Procedure-To 5 cc. of whole oxalated blood add 35 cc. of water, then 10 to 20 mg. of the urease powder, and shake. The estimated amount may be picked up on a knife blade. Let stand at room temperature, which must not be below 20”, for 20 minutes. Next add 5 cc. of the sodium tungstate solution, mix, and then 5 cc. of the 0.66 N sulfuric acid. Shake vigorously, let stand 10 minutes, and filter through a good grade of qualitative paper. Whatman No. 2 is satisfactory. It is not advisable to use the double acid-washed papers used in quantitative analysis, since all of these have been found to give appreciable ammonia blanks, while the qualitative papers give none. After filtration has proceeded for about 10 minutes, pour the filtrate back onto the paper. The first portion of the filtrate has been found to give greater light absorption when measured with a blue filter and also higher values for urea N and non-protein N.

Place 5 cc. of filtrate in a test-tube graduated at 20 and 25 cc. In a similar tube place 5 cc. of the working nitrogen standard, containing 0.075 mg. of nitrogen, and to each add distilled water to the 20 cc. mark. Mix by shaking.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


Prepare the nesslerizing solution by mixing 1 part of the gluconate and 1 part of the persulfate solutions. Pour this mixture into an equal volume of Nessler’s reagent (1 gluconate + 1 persulfate + 2 Nessler’s reagent). This mixture must be used within 15 minutes. To the standard and to ea& unknown add 4 cc. of the mixed reagent, dilute at once to the 25 cc. marq stopper with paraffined corks, and mix by vigorous shaking. Allow to stand for 15 minutes to develop full color, then compare in a calorimeter in the usual manner, using the proper filter. Comparisons should be com- pleted within 1 hour after nesslerization. The tubes should be kept stop- pered as much of the time as possible.

If a photoelectric calorimeter is used, the proper filter is inserted, full scale deflection secured with a distilled water blank, and the standard lea<;- then the unknowns. It is better thus to compare unknowns with ,; simultaneously prepared standard than it is to use a calibration curve, since the color varies greatly with temperature as shown by Pincussen (22), and with the age and lot of the reagents.

Calculation-When a visual calorimeter is used

Reading of standard

Reading of unknown X 15 = mg. urea N per 100 cc. blood

For direct reading photoelectric calorimeters with logarithmic scales the calculation is

Reading of unknown

Reading of standard X 15 = mg. urea N per 100 cc. blood

With calorimeters reading per cent of transmission only, the density must be calculated. This is defined as the log of (l/transmission) or log 1 - log T. Since scale readings are in terms of per cent of transmission, the above becomes log 100 - log of reading, or 2 - log of reading = density (D). Values of D for standard and unknown are then substituted in the formula

D unknown D standard X 15 = mg. urea N per 100 cc. blood

Test of Method-Results obtained by this method are shown in Tables I to III. In Table I is compared the stability of color with time for two specimens treated by four methods: (1) by the Wrenn method of direct nesslerization, (2) by the Looney method with gum ghatti, (3) by the method herein described, and (4) by the new distillation method followed by nesslerization with the new mixed reagent.

Comparison of results obtained by the last two methods shows absolute agreement for a period of 90 minutes in the first specimen and for 3 hours


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

C. J. GENTZKOW 541

in the second. The apparent increase in urea nitrogen shown in the first two columns is due to increasing turbidity. After the precipitate settles out, the color is actually lighter in intensity than it was originally. The Toduction in color intensity by the use of gum ghatti is also shown. All (<adings in Table I were made with the Evelyn calorimeter with a No. 490

TABLE I

Rate of Change in Color of Two Hydrolyzed Blood Filtrates Directly Nesslerized Compared with Distillation and Nesslerization of the Same Specimen

All values are expressed as mg. of urea N per 100 cc. of blood. - lessler’s reagenf

only* Time Nessler’s reagent + gum ghattit

.~ _ 0 min. 16.7 5 “ 17.3

10 ‘l 17.7 15 (‘ 18.2 30 “ 18.5 45 “ 19.0 60 “ 19.7 90 “ 20.3

2 hrs. 23.6 3 I‘ Heavy ppt. 0 min. 17.2 5 “ 18.15

10 I‘ 18.7 15 (‘ 19.2 30 “ 20.4 45 “ 21.4 60 I‘ 23.2 90 “ 26.5

2 hrs. ‘Ieavy ppt. 3 “ “ I‘ 4 Ii I‘ ‘I

24 “ “ “

15.0 15.9 16.3 16.4 16.7 17.4 18.1 19.0 20.8

Heavy ppt. 15.4 16.6 17.0 17.2 18.2 19.7 20.6 22.8

Heavy ppt. ‘I ‘I ‘I I‘ “ I‘

* Nesslerized only; similar to the method of Wrenn. t Method of Looney.

-

New reagent Distilled

specimen + new reagent

16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 17.2 16.7 19 .O (Clear) 16.7 16.3 17.0 17.15 17.1 17.15 17.1 17.15 17.1 17.1 17.1 17.1 17.1 17.1 17.1 17.1 17.1 17.1 17.1 17.1 17.1 17.45 17.1 18.6 (Clear) 17.1

filter. All distillations reported in this and in Table II were in duplicate, and blanks and standard nitrogen solutions were distilled as controls.

In Table II are shown a number of determinations on actual specimens from hospitalized patients. These determinations were made by the distillation method and the results compared with those obtained by the new method of direct nesslerization with the photoelectric calorimeter. They were compared also with the results obtained with a visual colorim- eter, both with and without the aid of a Wratten filter No. 75.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


Values determined by the method of direct nesslerization agree closely with those by the method requiring initial distillation of the ammonia when read on the photoelectric calorimeter. The maximum difference in the determinations listed is 2.2 per cent and the average 1.15 per cent. Results obtained when the readings were made in a visual calorimeter supplied

TABLE II

Comparative Values for Urea Nitrogen by New Distillation and New Direct Nessleri- zation Method; Readings Made by Photoelectric and Visual Calorimeters

All values in mg. per 100 cc. of whole blood.

New distillation method (Evelyn

colori;~~4$ filter T Visual calorimeter Xvelyn calorimeter,

filter No. 490

-

1

-

Per cent deviation I- With W&ten filter No. 75 Without filter

f2.0 13.2

-0.4 16.5

ZkO 16.8

-1.3 +2.1 14.7 12.0 +1.4 +1.5 13.0 12.2 -1.0 21.0 f0 21.5 19.0 +1.9 15.8 14.3 -0.9 77.4 +1.1 -1.2 +0.5 +2.2 9.4 +0.9 5.8 i’Jo reading

possible -

* Case of toxemia of pregnancy; no reading possible by visual calorimeter without filter because of difference in color quality.

13.23 13.25 16.26 16.25 16.7 16.7 15.6 14.5 14.6 13.2 20.5 21.0 16.0 76.0 43.7 17.2 18.0

9.1 5.52* 5.51

13.5

16.2

16.7

15.4 14.8 14.8 13.4 20.3 21.0 16.3 75.3 44.2 17.0 18.1

9.3 5.56

Direct nesslerization method

with the Wratten filter are also in reasonable agreement. Without the filter there is considerable error.

The one instance of extremely low urea nitrogen concentration was a case of suspected toxemia of pregnancy. It was absolutely impossible to do a determination on this specimen by any of the older methods, for upon the addition of Nessler’s reagent a very heavy precipitate immediately formed.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

C. J. GENTZKOW 543

As a final check on the method, experiments were conducted to determine the extent of recovery of urea added to blood samples, the urea concentration of which had already been determined. These data are recorded in Table III. Recoveries ranged from 97.2 to 98.6 per cent.

Precautions-A few precautions must be observed in using this method. The range over which good proportionality between standard and unknown holds is about a-fold. With a standard equal to 15 mg. of urea nitrogen per 100 cc., the concentration of the unknowns should be not less than 8 nor more than 30 mg. per 100 cc. For ranges beyond these limits a new standard must be prepared, or the filtrate diluted before nesslerization.

The clearest blood filtrates are produced by the original method of Folin and Wu. It is superior to any of the modifications, particularly those in which the acid is added before the tungstate or in which blood is added directly to ~/12 acid. Filtrates prepared in such manner show an appreciable absorption of light when measured in a photoelectric calorimeter with

TABLE III

Recovery of Urea Nitrogen Added to Blood

All values in mg. per 100 cc. of whole blood.

Urea N originally present

12.8 15.3 10.3 15.6

Amount added Amount recovered Per cent recovered -.

5.0 17.5 98.0 15.0 29.6 97.7 25.0 34.8 98.6 45.0 58.9 97.2

a blue filter. This absorption varies from 8 to 12 per cent as compared with 2 per cent by the original Folin-Wu filtrate. The zinc hydroxide filtrate of Somogyi (23) offers no advantage.

It is essential that the first portion of the filtrate be poured back onto the filter, for this first portion not only shows less transparency when measured with a blue filter but also yields higher values for non-protein and urea nitrogen.

Although urease solutions are frequently used, and their activity may be preserved indefinitely by the use of glycerol, such solutions nevertheless tend to develop ammonia blanks, even when relatively fresh, owing to the action of a secondary enzyme as shown by Howell (24). This difficulty is avoided by using a dry preparation. The double strength urease powder prepared by Squibb has a sufficiently high enzymatic activity and is entirely free of ammonia and other non-protein nitrogen. These criteria were not met by the other commerical preparations tested.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


SUMMARY

A simple accurate method of determining urea nitrogen in blood by direct nesslerization is described. No elaborate apparatus nor unusual reagents are required.

In the method potassium persulfate is used to prevent the reduction of Nessler’s reagent and potassium gluconate to restrain the oxidizing action of the persulfate, form a soluble mercury complex, or to do both. By these means a color is produced which is stable for a minimum of 1 hour. The final reaction mrxture remains clear and sparkling for days.

Spectrophotometric study of the color of the final nesslerized solution indicates that color comparisons should be made between the wave-lengths of 490 and 510 rnp with 500 rnp the optimum point. Appropriate filters for photoelectric and visual calorimeters are described.

As a basis for accurate comparison a new distillation procedure is out- lined which gives accurate results with the amount of urea normally present in 0.5 cc. of whole blood.

BIBLIOGRAPHY

1. Van Slyke, D. D., and Cullen, G. E., J. Biol. Chem., 24, 117 (1916). 2. Van Slyke, D. D., J. Biol. Chem., 73, 695 (1927). 3. Karr, W. G., J. Lab. and Clin. Med., 9, 329 (1924). 4. Barrett, J. F., Biochem. J., 29, 2442 (1935). 5. Raices, A. E., Rev. m&Z-q&. patol. femenina, 6, 531 (1935); Chem. Abst., 30,

126 (1936). 6. Wrenn, H. T., J. Lab. and Clin. Med., 22, 1040 (1937). 7. Taylor, W. F., Hayes, E. R., and Wells, B. B., J. Lab. and CZin. Med., 23, 188

(1937). 8. Hawk, F. L., and Andes, J. E., Am. J. CZin. Path., Tech. Suppl., 2, 153 (1938). 9. Naumann, H. N., J. Lab. and CZin. Med., 26, 405 (1940).

10. Hoffman, W. S., and Osgood, B., J. Lab. and CZin. Med., 26, 862 (1940). 11. Ohllson, W., Acta physiol. Stand., 1, 278 (1940); Chem. Abst., 36, 4053 (1941). 12. Sure, B., and Wilder, E. M., J. Lab. and CZin. Med., 26, 874 (1941). 13. Hughes, J., and Saifer, A., J. Lab. and CZin. Med., 27, 391 (1941). 14. Folin, O., Laboratory manual of biological chemistry, New York, 5th edition,

275 (1934). 15. Looney, J. M., J. BioZ. Chem., 66, 189 (1930). 16. Folin, O., and Svedberg, A., J. BioZ. Chem., 66, 77 (1930). 17. Miller, H. S., Ind. and Eng. Chem., Anal. Ed., 8, 50 (1936). 18. Kemmerer, G., and Hallett, L. T., Ind. and Eng. Chem., 19, 1295 (1927). 19. Redemann, C. E., Ind. and Eng. Chem., Anal. Ed., 11, 635 (1939). 20. Folin, O., and Wu, H., J. BioZ. Chem., 38, 81 (1919). 21. Koch, F. C., and McMeekin, T. L., J. Am. Chem. Sot., 46, 2066 (1924). 22. Pincussen, L., J. Lab. and CZin. Med., 26, 1062 (1941). 23. Somogyi, M., J. BioZ. Chem., 86, 655 (1930). 24. Howell, S. F., J. BioZ. Chem., 129, 641 (1939).


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

John M. MasenCleon J. Gentzkow and With the assistance of

NESSLERIZATIONNITROGEN BY DIRECT

DETERMINATION OF BLOOD UREA AN ACCURATE METHOD FOR THE

1942, 143:531-544.J. Biol. Chem.

http://www.jbc.org/content/143/2/531.citation

Access the most updated version of this article at

Alerts:

When a correction for this article is posted•

When this article is cited•

alerts to choose from all of JBC's e-mailClick here

tml#ref-list-1

http://www.jbc.org/content/143/2/531.citation.full.haccessed free atThis article cites 0 references, 0 of which can be


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/content/143/2/531.citation

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;143/2/531&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/143/2/531.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=143/2/531&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/143/2/531.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/143/2/531.citation.full.html#ref-list-1

http://www.jbc.org/content/143/2/531.citation.full.html#ref-list-1

http://www.jbc.org/

AN ACCURATE METHOD FOR THE DETERMINATION OF BLOOD … · an accurate method for the determination...

Documents

Transcript of AN ACCURATE METHOD FOR THE DETERMINATION OF BLOOD … · an accurate method for the determination...